Rotaviruses are the single most important etiologic agents of severe diarrhea of infants and young children worldwide. Because only four human rotavirus serotypes appeared to be epidemiologically important we developed a quadrivalent rotavirus vaccine containing representatives of each of the 4 serotypes: rhesus rotavirus (RRV) a VP7 serotype 3 strain (the Jennerian approach), and three human RV-RRV reassortants, each possessing ten RRV genes and a single human RV gene that encodes VP7 (a major outer shell protein) that is responsible for serotype 1, 2, or 4 specificity (the modified Jennerian approach). Four randomized, placebo-controlled, coded, efficacy trials of this orally delivered vaccine used at a dose of 4x105 PFU have been completed in 5958 infants and young children, (2 in the U.S. [783 and 695 children] and one each in Finland [2273 children], and Venezuela [2207 children]). This quadrivalent formulation, given orally in three doses, was highly effective in preventing severe rotavirus diarrhea with efficacy ranging from 69% to 91%. The vaccine was 75% to 100% effective in preventing dehydrating diarrheal illnesses. Because of the demonstrated safety (transient fever) and efficacy of the quadrivalent vaccine, the Advisory Committee on Immunization Practices in 1998 recommended its use for infants at 2, 4, and 6 months of age. Subsequently, in August 1998 the FDA granted a Biologics License to Wyeth Laboratories. However, after an estimated 1.5 million doses had been given to about 1 million infants, the CDC in July 1999 recommended postponing further vaccination because of the occurrence of 15 cases of intussusception following vaccination. Although the number of cases reported was within the expected value (considering a background rate of 51 per 100,000 [or about 1 case per 2000 infants during the first year of life in data from N.Y. State]), it was of concern that 11 of the 15 cases occurred within 1 week of receiving the first dose of vaccine. Subsequently in October 1999 the ACIP withdrew its previous recommendation because of additional data extending the initial findings of a link with intussusception. In conjunction with these events, Wyeth-Lederle withdrew the vaccine. In other studies, we evaluated collaboratively the feasibility of administering the licensed rhesus rotavirus-based vaccine to neonates to determine its reactogenicity and immunogenicity with different administration schedules. We also evaluated collaboratively several second and third generation vaccines: (i) a quadrivalent reassortant rotavirus vaccine that possesses a single VP7 gene from a human strain with serotype 1, 2, 3, or 4 specificity and the remaining ten genes from bovine RV (UK); (ii) a reassortant vaccine (Wa x UK) that contains a single gene from the human Wa strain, that encodes VP4:1A specificity, in a background of 10 bovine (UK) genes; (iii) a reassortant candidate vaccine Wa x (DS-1 x UK) that possesses the VP4 gene from the Wa strain (VP4:1A) as well as the VP7 gene from human strain DS-1 that encodes G2 specificity, in a background of 9 bovine (UK) genes; (iv) and a 30?C cold-adapted, temperature-sensitive, VP7:1 strain (D[75-82]) (see Hoshino and Ishida reports).